Data scientists are one of the most sought-after professions today because they can extract valuable information from large volumes of data in their research. To do this, they must be able to master programming languages developed explicitly for data science. However, these languages often contain a large number of errors which can then bias the research results. Detecting and eliminating these errors is the work of Pierre Donat-Bouillud, Ph.D., from the Programming Research Laboratory at FIT CTU, who received a grant from the European Commission for his research.
Using programming languages, a data scientist can perform predictive modelling in a particular field, in other words, creates algorithms capable of predicting future developments. However, errors, so-called data bugs, occur during programming. They can cause incorrect calculations or research results. Pierre Donat-Bouillud from the Programming Research Laboratory at FIT CTU works on detecting these errors in the “DataBugs: Finding Bugs in Data Science Codes” project.
“A data error is a bug or problem in a computer programme or system that can cause problems such as data loss, system crashes or miscalculations,” Pierre Donat-Bouillud described the issue, adding: “In my research, I look for these errors in the codes of programming languages for data science. The research aims to develop new programming language technologies to help data scientists and use the R programming language to validate their new ideas empirically.”
In his research, he develops a project for which he has already received an international award called Seal of Excellence under the Horizon 2020 call. The Seal of Excellence award is a “certificate of quality” awarded to projects that have crossed the eligibility threshold for funding but have not been funded. Based on this certificate, the scientist from FIT CTU received a grant funded by the Johannes Amos Comenius Operational Programme. That is under the responsibility of the Ministry of Education, Youth and Sports and is co-financed by the European Union.
Pierre Donat-Bouillud works at FIT CTU not only as a scientist but also as a teacher. He has opened a new course, Advanced Program Testing, for master’s students, which aims to introduce advanced techniques for testing programmes beyond writing unit tests. Pierre was a PhD student at the Sorbonne University in Paris, and his forte is programming languages.